A nanobody-based tri-specific NK cell engager targeting CD5 triggers antitumor immunity

Abstract

The poor prognosis of patients with recurrent or refractory T cell malignancies emphasizes the need for improved immunotherapies. CD5 is a characteristic marker of malignant T cells and is expressed on almost all normal T cells. Therefore, for treating T cell malignancies, focusing on natural killer (NK) cells lacking CD5 expression may elicit a better safety profile than that by T cell-based therapies. We generate a CD5-targeted NK cell engager (NKCE) through the specific binding of CD16a nanobody and a high-affinity anti-CD5 antibody. Its antitumor potency is demonstrated in vitro. After incorporating interleukin (IL)-15Rα/IL-15, the modified tri-NKCE exhibits stronger antitumor efficacy against CD5⁺ malignant tumor cells, with the production of more cytokines and chemokines. In vivo, tri-NKCE exhibits stronger cytotoxicity by enhancing NK cell proliferation. Compared with chimeric antigen receptor (CAR)-T cells, this tri-NKCE exhibits no toxicity to normal T cells. In conclusion, tri-NKCE offers a safer and cost-effective immunotherapy against T cell malignancies.

Keywords: CD16a; CD5; T cell malignancies; immunotherapy; nanobody; natural killer cells.

Graphical abstract

Figure 1

Identification and characterization of anti-CD5 scFvs and anti-CD16a nanobody (A) Flow cytometry analysis shows that the 1H6 and 2D7 can bind to CD5⁺ CCRF-CEM, MOLT-4, and Jurkat cells but not bind to CD5⁻ Raji cells. (B) Binding specificity of 1H6 and 2D7 to CD5-ECD protein was assessed by ELISA. (C) Binding affinity of 1H6 for CD5-ECD protein was determined by SPR. The K_D value is shown. (D) Schematic illustration of library construction and screening of nanobodies by phage display technology. This graph was created with BioRender.com. (E) Phage ELISA results showing the binding specificity of P60 to CD16a (158V/F)-ECD proteins. (F) Binding affinity of P60 to CD16a (158V/F)-ECD proteins was determined by BLI. K_D values are shown. (G) Binding capacity of P60 to HeLa-CD16a (158V), HeLa-CD16b (NA1), and HeLa cells was detected by IF assay. Scale bar, 20 μm. (B and E) Data are presented as mean ± SD (n = 3 technical replicates). NC, negative control. See also Figures S1 and S2.

Figure 2

In vitro activity of 1H6-NKCE (A) Schematic illustration of 1H6-NKCE structure. (B) Binding affinity of 1H6-NKCE and 1H6 to CD5 was analyzed in a series of dilution concentrations on ELISA plates coated with the indicated concentration of CD5-ECD protein. (C) Binding affinity of 1H6-NKCE and P60 to CD16a was analyzed in a series of dilution concentrations on ELISA plates coated with the indicated concentration of the CD16a (158V)-ECD protein. (B and C) NC, negative control. Data are presented as mean ± SD (n = 3 for technical replicates). (D) Flow cytometry analysis showing the phenotype of the purified NK cells and surface expression of CD16. (E) Binding capacity of 1H6-NKCE to NK cells via CD16a was detected using flow cytometry. (F) Cytotoxicity of NK cells mediated by 1H6-NKCE, 1H6, or P60 at various concentrations was determined by luciferase-based cytotoxicity assays (E:T = 1:1 ratio, two-way ANOVA followed by Tukey’s post hoc test). (G) Cytotoxicity of NK cells mediated by no antibody, 1H6-NKCE, 1H6, or P60 against HeLa-CD5 cells was assessed by the RTCA assays (E:T = 1:1 ratio, 60 h). TC, only target cells. (H) Representative flow cytometry analysis showing the percentages of CD107a-expressing NK cells in the presence or absence of CCRF-CEM cells. (I) Quantification and statistical analysis of the data in (H). Statistical analysis was determined using two-way ANOVA followed by Dunnett’s post hoc test. (F, G, and I) Data are presented as mean ± SD (n = 3 for technical replicates from one donor). See also Figures S3–S6.

Figure 3

1H6-15-NKCE enhances NK cell function against CD5⁺ target cells and primary hematologic malignant cells ex vivo (A) Schematic illustration of 1H6-15-NKCE containing the IL-15Rα/IL-15 complex. (B) Percentages of residual CD56⁻ target tumor cells were determined by flow cytometry. (C) Comparison of NK cell cytotoxicity mediated by IL-15-Fc, 1H6-NKCE, and 1H6-15-NKCE at various concentrations against CCRF-CEM cells was performed by luciferase-based cytotoxicity assays (two-way ANOVA followed by Dunnett’s post hoc test). (D) Direct lysis of PBMCs against target tumor cells mediated by no antibody, IL-15-Fc, 1H6-NKCE, or 1H6-15-NKCE (two-way ANOVA followed by Tukey’s post hoc test). (E) Tumor lysis of NK cells against HeLa-CD5 cells mediated by no antibody, IL-15-Fc, 1H6-NKCE, or 1H6-15-NKCE was detected by the RTCA assays. TC, only target cells. (F) Antibody-mediated NK-NK lysis was assessed using 4 h calcein-release cytotoxicity assays. (G) Flow cytometry analysis showing the quantified mean fluorescence intensity (MFI) of CD5 in BMMNCs or PBMCs isolated from patients (one-way ANOVA followed by Dunnett’s post hoc test). NC, negative control. (H) Representative flow cytometry analysis showing the percentages of residual primary tumor cells. (I) Quantification and statistical analysis of data in (H). (B, C, E, and G) Data are presented as mean ± SD (n = 3 for technical replicates from one donor). (D, F, and I) Data are presented as mean ± SD for 3 donors. (B and I) Statistical analysis was determined using one-way ANOVA followed by Tukey’s post hoc test. See also Figures S7–S11.

Figure 4

Transcriptomic profile and cytokine secretion were associated with stronger responses of NK cells mediated by 1H6-15-NKCE (A) Venn diagram representing the number of DEGs significantly regulated by each of indicated antibodies compared to the control group (no antibody treatment). (B) Volcano plot representing the analysis of differential expression patterns between NK cells activated with 1H6-15-NKCE and those treated with 1H6-NKCE. (C) Top GO terms of biological processes enriched in DEGs between 1H6 and 15-NKCE-activated NK cells and 1H6-NKCE-activated NK cells. (D) GSEA analysis identifying upregulated pathways associated with NK cell-mediated antitumor functions in 1H6-15-NKCE treatment relative to 1H6-NKCE treatment. (E) Unsupervised clustering analysis showing the expression level of NK cell-mediated cytotoxic pathway-related genes across all clusters. (F–H) ELISA analysis of TNF-α, IFN-γ, and GZMB levels on supernatants harvested from untreated or antibody-treated NK cells co-incubated with target tumor cells (one-way ANOVA followed by Tukey’s post hoc test). (I) Representative flow cytometry analysis showing the surface CD69 expression in NK cells from PBMCs after the treatment with no antibody, 1H6-NKCE, or 1H6-15-NKCE. The percentages of CD69-expressing NK cells are shown in the presence or absence of CCRF-CEM cells. (J) Quantification and statistical analysis of the data in (I). (K) Degradation marker CD107a was measured in NK cells by flow cytometry in the presence of target cells with various doses of 1H6-NKCE or 1H6-15-NKCE treatment. Data are presented as mean ± SD for 4 donors. (F–H and J) Each dot represents the data from one donor, and mean values ± SD for 4 donors are shown. (J and K) Statistical significance was determined using two-way ANOVA followed by Tukey’s post hoc test. TC, target cells. See also Figure S12.

Figure 5

1H6-15-NKCE induces potent NK cell survival and proliferation (A) Top GO terms of biological processes enriched in DEGs between 1H6-NKCE treatment versus 1H6-15-NKCE treatment (n = 3 donors). (B) Heatmap of selected upregulated genes based on cell proliferation function. (C) GSEA analysis identifying upregulated pathways related to cell proliferation in 1H6-15-NKCE treatment relative to 1H6-NKCE treatment. (D) The representative CFSE dilution profile showing the proliferation of NK cells induced by different treatment groups. Control represents the fluorescence intensity of CFSE-labeled NK cells with no antibody treatment on the first day. (E) Pooled analysis depicts the proliferation percentage of NK cells. (F) Representative histograms showing the cell death of different treatments via live/dead near-IR (infrared) staining. (G) Statistical analysis of pooled cell viability. (E and G) Each dot represents the data from one donor, and mean values ± SD for 4 donors are shown (one-way ANOVA followed by Tukey’s post hoc test). See also Figures S13–S15.

Figure 6

1H6-15-NKCE exhibits superior antitumor activity in a in vivo CCRF-CEM xenograft model (A) Schematic illustration of in vivo experiments using luciferase-expressing cells in a mouse xenograft model treated with NK cells activated by antibody and cytokine administration. This diagram was created with BioRender.com. (B) Tumor burden for each group was monitored by IVIS imaging (n = 5 mice per group). The untreated (UT) group received CCRF-CEM-ffLuc cells only. (C) Bioluminescent imaging data for each group was monitored for 25 days after immune effector cells infusion. Data are presented as mean ± SD (two-way ANOVA followed by Tukey’s post hoc test). (D) Kaplan-Meier curves representing the percent survival of the experimental groups. Statistical significance was determined using a log rank (Mantel-Cox) test. (E) Assessment of NK cell numbers in peripheral blood. Blood draws were taken on days 7, 14, and 21 (post-NK infusion). (F) Quantification of the percentage of CD45⁺CD3⁻CD56⁺ NK cell population in spleens after mice were sacrificed at day 21. (E and F) Each dot represents the data from a mouse, and mean values ± SD for 5 mice are shown (one-way ANOVA followed by Tukey’s post hoc test). See also Figure S16.

Figure 7

Comparison of antitumor activity of anti-CD5-CAR-T cells to that of NK cells activated by 1H6-15-NKCE in vitro and in vivo (A) Comparison of the cytotoxicity of 1H6-15-NKCE-activated NK cells and anti-CD5-CAR-T cells against CD5⁺ tumor cells was performed by luciferase-based cytotoxicity assays. (B) Direct cytotoxicity of 1H6-15-NKCE-activated NK cells and anti-CD5-CAR-T cells against HeLa-CD5 cells was assessed by RTCA assays (E:T = 2:1 ratio, 45 h). TC, target cells. (C) On-target off-tumor cytotoxicity of anti-CD5-CAR-T cells, MOCK-T cells, NK cells, or 1H6-15-NKCE-activated NK cells against autologous T cells was determined by luciferase-based cytotoxicity assays (one-way ANOVA followed by Tukey’s post hoc test). (D) Tumor burden was monitored by IVIS imaging (n = 5 mice per group). The untreated (UT) group received MOLT-4-ffLuc cells only. (E) Bioluminescent imaging data of each group were monitored after effector cells infusion. (F) Body weight curve of each group at different days. (G) Kaplan-Meier survival curves of five groups. Statistical significance was determined using a log rank (Mantel-Cox) test. (H) Levels of IFN-γ and IL-6 were measured by ELISA on mice serum collected from peripheral blood on different days after immune effector cells infusion. (A and C) Data are presented as mean ± SD (n = 3 replicates from one donor). (E, F, and H) Data are presented as mean ± SD (n = 5 mice). (A, E, F, and H) Statistical significance was determined by two-way ANOVA followed by Tukey’s post hoc test. See also Figures S17–S20.